What difference in AI units consumption can be expected when using AI in the Cloud compared to Suite?

When using AI in the Cloud as opposed to on-premise solutions offered in the Suite, the consumption of AI units can be expected to differ due to several factors, such as resource allocation, scalability, and performance optimization by the AI service provider.

Cloud services often operate on a consumption-based pricing model, which might seem less efficient if there is a standard ongoing demand. In such cases, the operational overhead and resource needs in the cloud environment—primarily driven by the necessity of ensuring uptime, addressing latency, and managing extensive data traffic—can lead to increased AI units consumption, which is responsible for a potential increase in costs.

Moreover, different cloud architectures and the services they provide can necessitate higher levels of resource availability compared to running AI models from a local suite. Hence, a +20% increase in consumption aligns with the additional resources needed to effectively manage AI workloads in a cloud infrastructure compared to Suite environments.

In light of these considerations, a +20% increase in AI unit consumption when using cloud services is a reasonable expectation.